Serum-Protein-Based Indices for the Progression of Fracture Healing and Nonunion

Aim 1 Learning Set: At the earliest time in the study when 10 non-unions have been identified, we will select 10 control patients matched for comorbidities (including a body mass index>40, smoking history, and diabetes) from the group that has been enrolled for comparison to the 10 non-union patients. The proteins showing the most substantial changes in response to either their initial or end point measurements in both healer and non-healer sets will be identified. Temporal clustering and biological assessment tools will be used to identify the biological processes and time frames showing altered serum levels for these markers. A selection process will then use a set of pair wise iterative processes to identify both single markers and multiple marker subsets that show the best correlations to both mRUST and union/nonunion diagnosis prior to the current time that the diagnosis for failed healing is made.

Aim2 Validation Set: As soon as the next group 10 non-union patients are identified a new group of 10 healers will be selected and the target group of protein identified in phase one will be tested for robustness in a second study with this new set of non-healers healers. Assessments of the robustness will be based on level of detection (LOD) and reproducibility of LOD and diagnostic performance. A third assessment will be carried out with the total group of 20 non healers identified in both the first and second phases of the project by rescreening this group against profiles of a final and novel set of 40 new healers. This new set of healers will be selected from the total patient sets of profiles that have been collected across the entire study period and again will be randomly selected and matched for comorbidities. During each subsequent screen we will refine the marker set further taking into account both performance features, qualitative factors related to a protein's biological functions and relationships to the processes of healing, and their correlation to the clinical parameters that we are using to assess progression of healing (union). The last stage of assessment provides an independent technical biochemical/immunological validation for the ten best performing proteins. This assessment will be made using a conventional Western blot assay across a randomized set of 20 profiles from both healers and non-healers who had been previously screened using the microarray format. This last assay is a quality control step for protein identification specification.